Author's personal copy - civil.iisc.ac.incivil.iisc.ac.in/~anbazhagan/pdf/Anbu IISc J-17.pdf · batore earthquake ofMW =6.2 occurred near the Bhavani fault that shook the city. In

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Multi-criteria seismic hazard evaluation for Bangalore city, India

P. Anbazhagan a,*, K.K.S. Thingbaijam b, S.K. Nath b, J.N. Narendara Kumar a, T.G. Sitharam a

a Department of Civil Engineering, Indian Institute of Science, Bangalore 560 012, Indiab Department of Geology and Geophysics, Indian Institute of Technology, Kharagpur 721 302, India

a r t i c l e i n f o

Article history:Received 5 May 2008Received in revised form 26 May 2009Accepted 4 January 2010

Keywords:Seismic microzonationPeak ground accelerationSeismic hazardSite responseHazard index

a b s t r a c t

Different seismic hazard components pertaining to Bangalore city, namely soil overburden thickness,effective shear-wave velocity, factor of safety against liquefaction potential, peak ground accelerationat the seismic bedrock, site response in terms of amplification factor, and the predominant frequency,has been individually evaluated. The overburden thickness distribution, predominantly in the range of5–10 m in the city, has been estimated through a sub-surface model from geotechnical bore-log data.The effective shear-wave velocity distribution, established through Multi-channel Analysis of SurfaceWave (MASW) survey and subsequent data interpretation through dispersion analysis, exhibits site classD (180–360 m/s), site class C (360–760 m/s), and site class B (760–1500 m/s) in compliance to theNational Earthquake Hazard Reduction Program (NEHRP) nomenclature. The peak ground accelerationhas been estimated through deterministic approach, based on the maximum credible earthquake ofMW = 5.1 assumed to be nucleating from the closest active seismic source (Mandya–Channapatna–Banga-lore Lineament). The 1-D site response factor, computed at each borehole through geotechnical analysisacross the study region, is seen to be ranging from around amplification of one to as high as four times.Correspondingly, the predominant frequency estimated from the Fourier spectrum is found to be pre-dominantly in range of 3.5–5.0 Hz. The soil liquefaction hazard assessment has been estimated in termsof factor of safety against liquefaction potential using standard penetration test data and the underlyingsoil properties that indicates 90% of the study region to be non-liquefiable. The spatial distributions of thedifferent hazard entities are placed on a GIS platform and subsequently, integrated through analyticalhierarchal process. The accomplished deterministic hazard map shows high hazard coverage in the wes-tern areas. The microzonation, thus, achieved is envisaged as a first-cut assessment of the site specifichazard in laying out a framework for higher order seismic microzonation as well as a useful decision sup-port tool in overall land-use planning, and hazard management.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Seismic hazard, in a broad perspective, refers to any kind ofphenomena related to earthquakes capable of imparting potentialdamages to the built and social environment. Although it is gener-ally defined as a specified level of ground shaking, several otherhazard entities such as landslides, liquefaction, tsunamis, seichesare often associated. The hazard is significantly controlled bychanges in geotechnical material properties during the earthquake.In fact, site-specific attributes related to surface geologic condi-tions can induce considerable alterations of the seismic motions(Aki, 1988; Field et al., 1992; Nath et al., 2000; Sitharam andAnbazhagan, 2008). It is, therefore, important to deliver appropri-

ate site-specific design ground motions for earthquake resistantstructural design and the hazard appraisal. In cognizance to theexistence of multiple hazard components, an appropriate decisionsupport tool for hazard classification would incorporate of everyaspect according to their likely contribution to the overall hazard.To that effect, seismic microzonation has been carried out throughmulti-criteria evaluation technique that accounts for several fac-tors such as site response, shear-wave velocity, landslide, geomor-phological features, besides the peak ground accelerations(Sitharam and Anbazhagan, 2008; Pal et al., 2008; Nath et al.,2008). Making improvements on the conventional regional hazardmaps, microzonation of a region predicts the hazard to much smal-ler scales (TC4-ISSMGE, 1999; Sitharam and Anbazhagan, 2008). Itinvolves subdivision of a region into individual areas having differ-ent potentials for hazardous earthquake effects, defining their spe-cific seismic behavior for engineering design, and land-useplanning (Anbazhagan and Sitharam, 2008b). The seismic microz-onation maps are generally envisaged to provide an effective tool

1367-9120/$ - see front matter � 2010 Elsevier Ltd. All rights reserved.doi:10.1016/j.jseaes.2010.01.001

* Corresponding author. Tel.: +91 8022932467; fax: +91 8023600404.E-mail addresses: [email protected], [email protected]

(P. Anbazhagan).

Journal of Asian Earth Sciences 38 (2010) 186–198

Contents lists available at ScienceDirect

Journal of Asian Earth Sciences

journal homepage: www.elsevier .com/locate / jseaes


for land-use planning, hazard mitigation and management, andstructural engineering applications, especially in vulnerable zonescharacterized with rapid urbanization and burgeoning population.

Bangalore metropolis area located at 12�580N, 77�360 coveringan area of about 220 km2, has been investigated for various seismichazard components, namely peak ground acceleration at bedrocklevel, site response, liquefaction potential, effective shear-wavevelocity, predominant frequency, and soil overburden thickness.Sitharam and Anbazhagan (2007a) estimated the peak groundacceleration distribution at bedrock level from synthetic groundmotions for a controlling seismic source in order to present adeterministic hazard scenario. Sitharam and Anbazhagan (2007b)evaluated site amplification factor, and predominant frequency ofsoil columns in the region from geotechnical borehole data usingSHAKE2000 (Ordonez, 2004). Anbazhagan and Sitharam (2008a)computed the average shear-wave velocity through Multi-channelAnalysis of Surface Wave (MASW) surveys at 58 locations acrossthe study region. The velocity measurements were, further,calibrated with those derived from the drilled boreholes data togenerate correlations between corrected standard penetration testN-values and shear-wave velocity (Anbazhagan and Sitharam,2009). Sitharam et al. (2007) investigated the overburden soil de-tails, and the soil liquefaction potential in terms of factor of safetyagainst liquefaction in the study region.

In the present study, the different hazard aspects are appraisedin order to establish their validity and usefulness, and eventually toprovide an amalgamation of the different factors in form of a haz-ard index map for the study region. The analysis, thus, carried outinvolved mapping the spatial distribution of these factors on a sin-gle reference system (1:20,000 scale resolution) using a Geograph-ical Information System (GIS) platform; each constituting athematic layer. Following a multi-criteria evaluation technique -analytical hierarchical process (AHP, Saaty, 1980, 1990), eachtheme and the features have been assigned weights and rankingsrespectively according to their perceived relative significances tothe seismic hazard. The layers are, thereafter, integrated throughspatial union to obtain the seismic microzonation map addressinga first-cut assessment of the site specific hazard to layout a frame-work for higher order seismic microzonation.

2. Study area and regional background

Bangalore city in the southwestern part of India come across asa vulnerable region with its expanding diverse huge populationbase, and extending urban infrastructure. The city is the principaladministrative, industrial, commercial, educational and culturalcapital of Karnataka State. It has been the fastest grown city andis presently ranked as fifth biggest city in India. Besides politicalactivities, Bangalore hosts several national scientific laboratories,defense establishments, small and large-scale industries. The cityemerged as ‘the silicon city of India’ with agglomeration of Infor-mation Technology corporate establishments, along with influx ofthousands of software professionals every year. The metropolisrepresents a booming commercial venue with expansive infra-structure and diverse population that continues to accommodatethe requirement of a modern urban setting. A study concerningthe seismological and geotechnical issues towards providing secu-rity for the inhabitants and safeguarding the infrastructural invest-ments is, therefore, not only significant but also essential.

Large number of earthquakes with different magnitudes has oc-curred often in this region (Bansal and Gupta, 1998). Recently,Ganesha Raj and Nijagunappa (2004) highlighted neo-tectonicactivities in the Karnataka region and suggested the current seis-mic zonation of Karnataka placed in lowest hazard zone i.e. zoneII of BIS hazard zonation code (BIS, 2002) to be inadequate. The re-gional environs of the city have been implicated with neo-tectonicand fault reactivation. A seismotectonic map depicting a circulararea of about 350 km radius around the city has been depicted inFig. 1. The active faults and lineaments along with associated seis-micity in the region have been examined in following up to theprevious studies of Dasgupta et al. (2000), and Ganesha Raj andNijagunappa (2004). The seismicity is accounted for with a MW

consistent earthquake catalogue covering a period of 200 yearsfrom 1807 to 2006. The seismic activities of the relevant faultsand implications to the fault-rupture patterns have been discussedin detail by Sitharam et al. (2006). The closest observable linea-ment extending SW–NE about 105 km with 5.2 km away fromthe city to the north Mandya–Channapatna–Bangalore Lineament.To the west of Bangalore, amongst several minor faults, Chikmag-

Fig. 1. A seismotectonic map of the study region depicting major linear tectonic features with the seismicity covering a period of 200 years: 1807–2006 (after Sitharam et al.,2006).

P. Anbazhagan et al. / Journal of Asian Earth Sciences 38 (2010) 186–198 187


alur fault to the north is seen trending in NW–SE orientation(Fig. 1) while Bhavali fault to the south traverses in slight variationfrom the NW-SE orientation. Southwest from the city, 1900 Coim-batore earthquake of MW = 6.2 occurred near the Bhavani fault thatshook the city. In the eastern Dharwar Craton region and closer toBangalore, Arkavati fault runs in similar trend to the Bhavali fault.Earthquakes of magnitude less than but close to MW = 5.0 havebeen observed close to Arkavati fault. To the north, the ChitradurgaBoundary Fault is seen trending in the N–S direction. A strike-slipmechanism is also indicated in the zone with Bangalore earth-quake of March 20, 1984, MW = 4.5 (Rastogi, 1992). On the otherhand, to the east, the faults are predominantly oriented in NE–SW fashion. Major faults include Kadiri Fault to the north; Main,Amirdi and Attur faults clustered to the southwest of the zone,and the Cauveri Fault traversing E–W on the southern boundary.Valdiya (1998) highlighted that the seismic activity is generallyconfined to linear belts related to crosscurrent and terrain-bound-ing faults, and shear zones implying possible reactivation of thePrecambrian faults.

Notably, there were over 150 lakes in and around the city. How-ever, most of them dried up due to soil erosions and encroach-ments for construction of residential/industrial buildings leavingonly 64 at present (Sitharam et al., 2006). The old lakes beds natu-rally have silty clay and sand that have been filled up with soil overwhich buildings/structures have been built.

3. Deterministic primary hazard distribution

Earthquakes primarily cause ground shaking that, at times, canbe devastating for buildings and subsequently the populace. Seis-mic hazard analysis, therefore, aims at the assessment of groundmotion potential at the site. Amongst several peak ground motionattributes (i.e. peak acceleration, peak velocity and peak displace-ment), peak ground acceleration (PGA) is often preferred as thehazard quantifier, which represents a short period ground motionparameter signifying damage potential to the buildings thus pro-viding an overall quantitative basis for the design codes and con-struction practices. However, period-specific spectral acceleration(SA) would also be a better index when specific buildings are con-sidered accounting for applicable resonance frequencies.

The seismic hazard assessment can be either deterministic orprobabilistic. The deterministic hazard analysis involves determi-nation of a controlling earthquake referred to as maximum credi-ble earthquake that produces the severest ground motion. Theassociated strong-motion parameters are estimated accordinglywithout considering probability of its occurrence. On the otherhand, the probabilistic analysis explicitly incorporates quantitativeuncertainties in the size, location, rate of recurrence and effects ofearthquakes. Both approaches are significant owing to the purpose,the seismic environment, and the scope of the assessment (McGu-ire, 2001; Anbazhagan et al., 2008). The deterministic approach isappropriate in cases of establishing a seismic framework towardsdisaster mitigation and management, and long-term earthquakehazard appraisal. Realistic seismic deterministic scenarios are alsouseful in case of seismic design and retrofits (Nath et al., 2009).However, in the cases where the nature of decision-making is tobe based on quantitative information involving uncertaintiesallowing constrains on investments according to the applicablescheme; for example, structural design requirements, financialplanning for earthquake losses (insurance), and investment forurbanization, a probabilistic approach would be more appropriate.

Sitharam and Anbazhagan (2007a) reported a deterministichazard analysis in the study region. Maximum credible earth-quakes corresponding to each tectonic feature have been com-puted through Wells and Coppersmith (1994) relation between

the length of sub-surface fault-rupture and magnitude, by consid-ering possible maximum fault-rupture in each case. The maximumfault-rupture dimensions are, however, constrained by the under-lying geology. The fault-rupture lengths, therefore, are construedfrom the observed seismicity, which indicated the pragmatic max-imum fault-rupture (segment) length to be 5% of the total faultlength. In order to establish the controlling seismic source, thePGA at the center of the Bangalore city were estimated using theregional attenuation relation of Iyengar and Raghukanth (2004)taking shortest distance from the different sources (faults and lin-eaments). The highest PGA of 0.159 g has been attributed for amaximum credible earthquake (MCE) of MW = 5.1 at Mandya–Channapatna– Bangalore Lineament, which exceeds more than50% of PGA due to other sources. The source has been, therefore,considered as the controlling seismic source for the deterministicassessment.

The stochastic simulation algorithm of Boore (1983) has beenemployed for the strong ground motion synthesis at borehole loca-tion points; about 620 bore holes have been dug for geotechnicalinvestigations (discussed in subsequent section) wherefrom thebasement depth information has been derived. MCE has beenplaced at a focal depth of 15 km since larger earthquakes (closerto MW = 5.0) in the source region are observed to be in the depthrange. The hypocenter distance from each borehole is taken to beshortest distance from the lineament. The average crustal shear-wave velocity has been taken to be 3.65 km/s. The geometric atten-uation G has been taken to be 1/R for R < 100 km and 1/10

pR for R

> 100 km (after Singh et al., 1999), in this R is hypocenter distancein km, the quality factor Q(f) has been taken to be equal to 488 f 0.88

(Sitharam et al., 2006; Sitharam and Anbazhagan, 2007a), f is thefrequency in Hz. A stress drop of 300 bars, and the high-frequencyband-limitation parameter, fmax, set to 35 Hz has been consideredto generate the strong ground motions at 80 Hz as suggestedappropriate for 15 km focal-depth earthquakes by Singh et al.(1999) for the Peninsular Indian region. A representative accelero-gram has been provided with Fig. 2a. The stochastic simulationsemploy random vibration theory to predict the peak ground accel-eration from time-domain simulations (Boore, 1983). Nevertheless,the number of simulations at closely located sites using the samesimulation parameters is not expected to yield highly fluctuatingresults, which otherwise would indicate huge uncertainty in themethod. The uncertainties in the source model and simulationparameters (more appropriate for probabilistic studies) have notbeen considered for the deterministic analysis (Nath et al., 2009;Anbazhagan et al., 2009)). To avoid minor fluctuations, specificranges of the predicted PGA values have been zoned through con-touring, as depicted in Fig. 2b, which exhibits a monotonic trendwith the highest value of 0.15 g to the northwest and lowest of0.10 g to the southeast.

4. Geotechnical attributes

The strong ground motions caused by an earthquake at a siteare greatly influenced by the underlying geotechnical propertiesof the local soil, implying the necessity to incorporate relevant geo-technical parameters in local specific seismic hazard assessment(Sitharam and Anbazhagan, 2008). Five parameters have been,therefore, considered in the present study: soil overburden thick-ness, effective shear-wave velocity, predominant frequency, siteresponse, and the factor of safety against liquefaction.

4.1. Soil overburden thickness and effective shear-wave velocity

The bedrock topography of a particular area illustrates the per-tinent soil overburden thickness which is an important geological

188 P. Anbazhagan et al. / Journal of Asian Earth Sciences 38 (2010) 186–198


parameter that significantly contributes to spatial variation of thestrong ground motion at the surface due to an earthquake. Sitha-ram et al. (2007) and Sitharam et al. (2008) developed a sub-sur-face model of the city using geotechnical bore-log data. A typicalbore log in the southern part of the study area is presented inFig. 3. A general depth-wise distribution of the soil type derivedfrom the bore-log data is given in Table 1. The data from 170 bore-holes covering the study region, as depicted in Fig. 4, have beenconsidered to estimate the corresponding soil overburden thick-ness. The spatial variation of the overburden thickness, obtainedthrough contouring using the estimated values at each borehole,is presented in Fig. 4. It is seen that the overburden thickness inthe study region varies from 1 m to 25 m but is predominantly inthe range of 5–10 m.

The shear-wave velocity, VS, distribution in the city has beenevaluated through multi-channel analysis of surface wave (MASW)survey and subsequent data analysis by Anbazhagan and Sitharam(2008a). The study covered 58 sites within the 220 sq. km of theurban center. Typical dispersion curve and shear-wave velocity isgiven in Fig. 5a and b. The recorded data have been selectedaccording to the associated highest signal-to-noise ratio. All tests

has been carried out with geophone interval of 1 m, source beingkept on both sides of the spread and source to the first and last re-ceiver were also varied from 5 m, 10 m and 15 m to avoid the ef-fects of near-field and far-field. The captured Rayleigh wave hasbeen analyzed using SurfSeis software.

The effective shear-wave velocity, VH, for the depth, d, of soil hasbeen computed as

VH ¼X

di

.X di

v i

� �ð1Þ

where di and vi denote the thickness (in m) and shear-wave velocityin ms�1 (at a shear strain level of 10�5 or less) of the ith formationor layer respectively, in a total of N layers. Shear-wave velocityaveraged over the upper 30 m (V30

S ) is accepted for site classificationpurpose as per National Earthquake Hazards Reduction Program(NEHRP, Building Seismic Safety Council, 2001), and InternationalBuilding code (IBC) classifications (Dobry et al., 2000; Kanli et al.,2006). Therefore, V30

S is generally considered for amplification andsite response studies. However, if the rock is found to be locatedwithin a depth of about 30 m, the effective shear-wave velocity of

Fig. 2. (a) A representative accelerogram generated through stochastic modeling at bedrock level, and (b) the deterministic spatial distribution of peak ground acceleration atbedrock level depicting a variation from 0.10 to 0.15 g in the Bangalore city.



soil thickness (overburden thickness), rather than V30S , is considered

since the presence of hard rock mass would implicate a higher V30S

(Anbazhagan and Sitharam, 2008c). The soil thicknesses were eval-

uated from the nearby borehole logs well within 500 m distance. Incompliance to the NEHRP nomenclature (Building Seismic SafetyCouncil, 2001), the site classification scheme has been adopted,

Fig. 3. An example of typical bore log from the study area.

Table 1General depth-wise distribution of the soil in the study region.

Layer Northwest Southwest Northeast Southeast

First Silty sand with clay, 0–3 m Silty sand with gravel, 0–1.7 m Clayey sand, 0–1.5 m Filled up soil, 0–1.5 mSecond Medium to dense silty sand, 3–6 m Clayey sand, 1.7–3.5 m Clayey sand with gravel, 1.5–4 m Silty clay, 1.5–4.5 mThird Weathered rock, 6–17 m Weathered Rock, 3.5–8.5 m Silty sand with Gravel, 4–15.5 m Sandy clay, 4.5–17.5 mFourth Hard rock, below 17 m Hard rock, below 8.5 m Weathered rock, 15.5–27.5 m Weathered rock, 17.5–38.5 mFifth Hard rock Hard rock Hard rock, below 27.5 m Hard rock, below 38.5 m



and the map, thus, obtained is depicted in Fig. 6. The ranges of V30S in

the region exhibits site class D (180–360 m/s), site class C (360–760 m/s), and site class B (760–1500 m/s) as per Anbazhaganet al. (2009).

4.2. Site Response and predominant frequency

The study region, predominantly, presents altered soil struc-tures due to dried up water bodies and reclamation of land(Anbazhagan and Sitharam 2009). The pertinent loose and silty soilconditions are prone to ground motion amplification that may beattributed to resonance effects of the velocity contrasts betweenthe unconsolidated overburden and bedrock, apart from rebound-ing effects within the stratigraphic layers. As seismic waves travelfrom bedrock to the surface, certain characteristics of the waves,such as amplitude and frequency content is modified as they passthrough the soil deposits, which in turn can transfer large acceler-ations to structures, particularly when the resulting seismic wavefrequency matches with the resonant frequencies of the structures.Therefore, high significance is associated to site specific ground re-sponse analysis to determine effect of local soil conditions onamplification of seismic waves, and hence the ground responsespectra for future design purposes. The response of a soil columnis generally seen to be dependent on the frequency of the base mo-tion as well as on the geometry and material properties of the soillayer above the bedrock (Anbazhagan et al., 2009).

Sitharam and Anbazhagan (2007b) performed a detailed groundresponse analysis in the study region. In this work, the sub-surfacemodel of the study area, represented by 170 geotechnical bore logsand 58 shear-wave velocity profiles, has been employed to accountfor the soil properties and synthetic ground motions at each bore-hole locations to compute the associated site effects through one-dimensional ground response with SHAKE2000 computer program(Ordonez, 2004). The program accounts for the non-linear soil

behavior by adopting equivalent linear steps using the shear mod-ulus and damping reductions curves. The values for modulus anddamping compatible with the effective strains in each geologicallayer are obtained through an iterative procedure. It is seen fromthe shear-wave velocity profiles that the hard rock/engineeringrock in the region is seen to be �760 m/s (site class B). The syn-thetic strong ground motion due to the controlling earthquake ofthe deterministic hazard analysis (Section 3), therefore, has beengiven as an input to the hard rock/engineering bedrock havingthe shear-wave velocity of 760 m/s. Accordingly, the peak acceler-ation values and acceleration time histories were computed at thetop of each sub layer. The site response has been quantified asamplification factor computed as a ratio of the peak horizontalacceleration at the ground surface to the peak horizontal accelera-tion at the bedrock. The latter is obtained from the synthetic accel-eration time history generated at each borehole.

Fig. 7a depicts the distribution of site amplification factor inthe study region. The site response distribution indicates predom-inantly higher site amplification (2–4 times) to the north, andlower site amplification (2 times) dominating most of the region,except for central and a few parts where site amplification is notobserved (i.e. 1 times). Noting that the overall soil overburdenthickness is rather shallow, the soil distribution that is mostlysand (silty and clayey) is associated with higher site amplifica-tion. To that extent, the predominant frequency is also seen highto the north. The site amplification, as the one observed in thecentral part of the study region, might be due to possible hardrock intrusions at the deeper sub-surface corroborated by the per-tinent borehole data, indicative of the geologic history of the area.Overall, the observed site amplification factors in view of the soilconditions and sediment thickness are consistent to those of sev-eral previous investigations.

The site predominant frequency is defined as the frequency ofseismic response of the soil columns corresponding to the maxi-

Fig. 4. The spatial distributions of overburden thickness of Bangalore city.



mum Fourier amplitude. It has been widely used to categorize thesoil in respect to the ground motion. Sitharam and Anbazhagan(2007b) estimated the predominant frequency distribution in theBangalore city from the Fourier spectrum computed usingSHAKE2000 at 170 boreholes sites. The predicted values have beencontoured and presented as a distribution map as depicted inFig. 7b. The predicted predominant frequency distribution is seento be predominantly in range of 3.5–9.5 Hz in the study region,while higher values are observed to the north of the region. Majorpart of the study area (Bangalore) have buildings with single to 3–4stories, which are prone to higher frequency (2–10 Hz) and thesevalues are within the predominant frequency distribution of thesoil in the study region.

4.3. Factor of safety against liquefaction

Soil liquefaction occurs when loose saturated unconsolidatedsoils transform from a solid state to liquefied state due to increas-ing pore water pressures, and thus decreasing effective stress, in-duced by their tendency to decrease in volume under drainedconditions when subjected to earthquake loading. Moderate to ma-jor earthquakes can cause liquefaction hazard. The hazard has been

seen more prone in loose to moderate granular soils with poordrainage, such as silty sands or sands and gravels capped or con-taining seams of impermeable sediments (Youd and Idriss, 2001).The attributing factors include the grain size distribution of soil,duration of earthquake, amplitude and frequency of shaking, dis-tance from epicenter, location of water table, cohesion and perme-ability of the soil.

In the study region, the liquefaction hazard assessment hasbeen carried out by Sitharam et al. (2007) using standard penetra-tion test (SPT) data and the underlying soil properties. The lique-faction susceptibility is a measure of an inherent resistance ofsoil to liquefaction, and can range from non susceptible, regardlessof seismic loading, to highly susceptible, which means that very lit-tle seismic energy is required to induce liquefaction. Liquefactionsusceptibility is evaluated based on the primary relevant soil prop-erties such as grain size, fine content, and density, degree of satu-ration, SPT-N values and age of the soil deposit in each of the borelogs. To that effect, the empirical assessment is decided on basisthat the soil is susceptible for liquefaction if (i) presence of sandlayers at depths less than 20 m, (ii) encountered water table depthless than 10 m, and (iii) SPT ‘N’ values indicating blow counts lessthan 20.

Fig. 5. Typical results from multi-channel analysis of surface wave (MASW) survey for the study area: (a) dispersion curve and (b) shear-wave velocity profile.



About 620 borehole data have been used to evaluate factor ofsafety against liquefaction in the terrain. The factor of safetyagainst liquefaction (FSL) of soil layers have been evaluated basedon the simplified procedure of Seed and Idriss (1971) and subse-quent revisions of Seed et al. (1983), Youd et al. (2001), and Cetinet al. (2004). To evaluate FSL, the earthquake induced loading is ex-pressed in terms of cyclic shear stress and compared to the lique-faction resistance of the soil. The former is represented by CyclicStress Ratio (CSR) while the latter by Cyclic Resistance Ratio(CRR). The surface consistent peak ground acceleration corre-sponding to the MCE is used to estimate the cyclic stress ratio(CSR). In the cyclic stress approach the pore pressure generationis related to the cyclic shear stress, hence the earthquake loadingis represented in terms of cyclic shear stresses. Liquefaction resis-tance of soil depends on how close the in situ state of soil is to thestate corresponding to failure, which have been assessed in situtest based on SPT ‘N’ values. Cyclic resistance ratio (CRR) has beenderived on corrected N value. The MCE of MW = of 5.1 necessitates amagnitude scaling factor (MSF, Seed and Idriss, 1982) to be evalu-ated as given below,

MSF ¼ 102:24

M2:56W

" #ð2Þ

The cyclic stress ratio caused by the earthquake is greater thanthe cyclic resistance ratio of in situ soil then liquefaction could oc-cur. FSL, is computed as follows,

FSL ¼ CRR7:5

CSR

� �MSF ð3Þ

The factor of safety for each layer of soil was arrived by consid-ering corresponding ‘‘(N1)60cs” values to arrive CRR and CSR forearthquake loading. The minimum factor of safety from each borelogs has been considered to represent the factor of safety against

liquefaction. Eventually, the FSL distribution for the city, which isdepicted in Fig. 8, have been mapped taking into account boththe empirical and geotechnical assessments. The reclaimed sitesof old lakes compiled through newspaper publications, accountsof the inhabitants, and old toposheets in a study conducted by In-dian Institute of Science, Bangalore has been indicated in the fig-ure. It is seen that a reclaimed site, namely Kurubarahalli Lake, inthe northeast of the study region exhibited high susceptibility tothe hazard, although most of the reclaimed sites do not. Only4.2% of the total area are seen to have FSL <1 whereas about14.7% have FSL between 1 and 2. About 12.5% have FSL between2 and 3, and about 68% have FSL >3. About 33% of the locationshave silty clay soil that may cause stress reduction in soil duringearthquake as they posses liquid limit >33 and plasticity index>12. The bore logs at the locations having FSL <1 indicate that veryloose silty sand with clay and sand is present up to a depth of about6 m which are classified as medium to fine sand with very low fieldSPT-N values. Shallow water tables (�1.2 m depth) are also foundin these locations. These factors may be attributed for the low fac-tor of safety. About 90% of the area in the study region have higherfactor of safety and are non-liquefiable. The city is, therefore, safefrom liquefaction hazard except at few locations where the over-burden is sandy silt complemented by shallow water table, not-withstanding that there has been no account of liquefaction inthe area.

5. Seismic microzonation

Multi-criteria assessment for hazard delineation leading to seis-mic microzonation has been accomplished previously in other In-dian regions – Guwahati City (Nath et al., 2007, 2008), SikkimHimalaya (Pal et al., 2008), and Delhi (Mohanty et al., 2007). Thehazard mapping is achieved through multi-criteria based decisionsupport tool formulated by Saaty (1980) referred to as Analytical

Fig. 6. NEHRP site classification map of Bangalore city (modified after Anbazhagan et al., 2009).



Hierarchal Process (AHP). The tool uses hierarchical structures torepresent a problem, and thereafter, develop priorities for thealternatives based on the consensus judgment. The technique uti-lizes organized priority in terms of weights assigned to each crite-ria or themes, which could be easily incorporated to the thematiclayers on a GIS platform. The weighting activities in multi-criteriadecision-making are effectively dealt with hierarchical structuring

and pair-wise comparisons wherein the judgment between twoparticular elements is formulated rather than prioritize an entirelist of elements. The process involves construction of a matrix ofpair-wise comparisons (ratios) between the factors of adoptedparameters depicting relative importance based on the assignedweightage. For example, six parameters are scaled as 1–6 on basisof the relative importance over one another such that 1 indicating

Fig. 7. The maps of Bangalore metropolis area depicting (a) site amplification, and (b) predominant frequency (PF) distribution respectively.



that the two factors are of equal importance, and 6 representingthat one factor is more important than the other, and correspond-ingly reciprocals of the scale (i.e., 1/1–1/6) show that one is lessimportant than others. The allocation of weights for the parameteror the theme depends on the relative importance of factors decidedthrough consensus opinion of participatory group of decision mak-ers. Accordingly, normalized weights are derived in each case.Within each theme, the values varies significantly and are henceclassified into various ranges or types collectively referred to asfeature of a thematic layer. The associated features are ranked orscored within the theme. The initial integral ranking, Ri, is normal-ized to ensure that no layer exerts an influence beyond its deter-mined weight using the following relation,

Rnorm ¼ ðRi � RminÞ=ðRmax � RminÞ ð4Þ

where Rnorm, Rmin and Rmax denotes the normalized, assigned mini-mum and maximum ranks respectively. In the present study, GISsoftware – ArcGIS 9.0 has been used for the purpose of thematicmapping through vector layer generation, and the spatial analysis.The advantage of GIS is the capability to store, manipulate, analyzeas well as display spatial and attribute data. The GIS framework alsoallows one to account for added levels of details and complexity,apart from facilitating easy querying.

The hazard themes, pertaining to the study region materializedas thematic layers on the GIS platform in the present analysis, are:(i) soil overburden thickness (SOT), (ii) effective shear-wave veloc-ity (ESV), (iii) factor of safety against liquefaction potential (FSL),(iv) peak ground acceleration (PGA) at seismic bedrock, (v) site re-sponse in terms of Amplification factor (SA), and (vi) predominantfrequency (PF). Each thematic layer has been georeferenced onUniversal Transverse Mercator coordinate system. The correspond-ing weights and the ranks to each thematic layer and the feature

ranks thereof are assigned accordingly to the apparent contribu-tion of the layers to the overall seismic hazard. The geological siteconditions greatly influences the strong ground motion at a siteand hence, higher importance is given to those attributes directlyconnected to geological and geotechnical site conditions (DST,2007, Sitharam and Anbazhagan, 2008). The basement topographyrepresented by SOT arbitrarily controls the site specific hazard,especially in the cognizance to the contrast in geophysical proper-ties between the basement and the soil deposits in the study re-gion, and hence has been accorded the highest weightage. Thesediment thickness implicates rebounding of the seismic wavesleading to site amplifications, and therefore, the ranks are decidedaccording to the increasing order of the thickness (or basementdepth) i.e. higher the depth, higher the rank. The next attribute isthe site class that is defined on basis of effective shear-wave veloc-ity. The soil liquefaction potential, evaluated in terms of FSL, isknown to be a determinant urban geotechnical hazard, especiallyat the reclaimed sites previously of natural water bodies, and

Fig. 8. The liquefaction hazard in the Bangalore city is characterized by the factor of safety against liquefaction. The site of old lakes (reclaimed land) are indicated with circleswith index: (1) Tumkur Lake, (2) Ramshetty Palya kere, (3) Oddarapalaya Lake, (4) Ketamaranahalli Lake, (5) Kurubarahalli Lake, (6) Agasana Lake, (7) Jakkarayana kere, (8)Dharmambudhi Lake, (9) Vijayanagar Chord Rd. Lake, (10) Marenahallli Lake, (11) Sampangi Lake, (12) Kalasipalya Lake, (13) Siddapura Lake, (14) Tyagarajanagar Lake, (15)Domlur Lake, and (16) Shule Lake. The existing lakes are referred to as tanks.

Table 2Pair-wise comparison matrix of themes and their normalized weights.

Themes SOT SC FSL PGA SA PF Normalized weights

SOT 6/6 6/5 6/4 6/3 6/2 6/1 0.2857SC 5/6 5/5 5/4 5/3 5/2 5/1 0.2381FSL 4/6 4/5 4/4 4/3 4/2 4/1 0.1905PGA 3/6 3/5 3/4 3/3 3/2 3/1 0.1429SA 2/6 2/5 2/4 2/3 2/2 2/1 0.0952PF 1/6 1/5 1/4 1/3 1/2 1/1 0.0476

SOT: soil overburden thickness; SC: site classification; FSL: factor of safety againstliquefaction potential; PGA: peak ground acceleration at the seismic bedrock; SR:site response in terms of maximum amplification factor; PF: predominantfrequency.



swampy tracts. The attribute is strongly connected to the loose soilconditions and is, therefore, accorded with the appropriate weigh-tage. Next in the order is PGA that constitutes the deterministichazard conforming to the basement. The 1-D site amplification fac-tor (SA) distribution essentially computed from the shear-wavevelocity profiles over the engineering bedrock comprises a factorthat adds up to the peak ground acceleration. Finally, PF is incorpo-

rated to address a generic hazard conditions for the building distri-bution. The building distribution in the study region comprisesmostly of 3–4 stories reinforced concrete structures that are gener-ally prone to high frequencies (2–10 Hz); well within the observedpredominant frequency range of the soil columns in the terrain (3–9.5 Hz). A generic scheme to address the overall higher raisedbuildings is, therefore, adopted in the present analysis with lowerpredominant frequency being assigned higher ranking in order tofacilitate an elementary hazard appraisal. The present framework,thus, accounts for the overall hazard considering the basin effect(soil overburden thickness, predominant frequency), soil effect(site classification, site response, liquefaction potential), and the ef-fect of the projected deterministic ground motion (peak groundacceleration). Table 2 presents the pair-wise comparison matrixfor the respective themes and their normalized weights. The nor-malized ranks assigned to the features of each theme are listedin Table 3. The thematic integration is achieved through the fol-lowing equation,

DHI ¼ ðSOTW � SOTR þ SCW � SCR þ FSLW � FSLR þ SOTW :SOTR

þ ESVW � ESVR þ DPGAW � DPGAR þ SAW � SAR þ PFW

� PFRÞ=X

W ð5Þ

where DHI represents the deterministic hazard index, and the sub-scripts – W and R have been assigned accordingly to indicate weightand ranking respectively.

The deterministic seismic microzonation map, achieved in thepresent analysis, is depicted in Fig. 9. The variations of DHI couldbe grouped into four classifications representing negligible, low,moderate, and high zones.

6. Discussion and conclusions

The present study focuses on delivering a seismic framework toenable decision on further investigations, especially, higher order

Table 3Normalized weights and ranks assigned to the respective themes and the featuresthereof for the thematic integration.

Themes Weight Feature Rank Normalized Rank

SOT (m) 0.2857 >20 5 1.0015–20 4 0.7510–15 3 0.505–10 2 0.2565.0 1 0.00

SC 0.2381 Site class D 3 1.00Site class C 2 0.50Site class B 1 0.00

FSL 0.1905 <1 3 1.001–2 2 0.50>2 1 0.00

DPGA (g) 0.1429 >0.15 5 1.000.14–0.15 4 0.750.13–0.14 3 0.50.12–0.13 2 0.2560.120 1 0.00

SA 0.0952 >4 4 1.003– 6 4 3 0.662– 6 3 2 0.331– 6 2 1 0.00

PF (Hz) 0.0476 63.5 5 1.003.5– 6 5.0 4 0.755– 6 7.5 3 0.507.5– 6 9.5 2 0.259.5– 6 11 1 0.00

Fig. 9. The deterministic seismic hazard microzonation map of Bangalore city; the listing of old lakes (reclaimed lands) have been given in caption of Fig. 8.



ward-specific analysis. This calls for a deterministic assessment,which usually preludes a probabilistic one. The study region is lo-cated in low seismicity zone. However, deadly earthquakes such asMW = 6.2 Latur 1993, MW = 5.8 Jabalpur 1997, and MW = 7.6 Gujarat2001 have occurred in the seismotectonic regime that encom-passes the peninsular India. The high uncertainty involved inassessment of the hazard predicates the significance of a determin-istic approach in the present study. A singular and dominant sce-nario has been assumed in the hazard projection that outstripsthose anticipated from other regional faults. Furthermore, higherpriority in the thematic integration process has been assigned tothe geotechnical themes, emphasizing more on the local geotech-nical hazard variations.

The final seismic hazard microzonation map obtained in thepresent analysis indicates that low hazard zone occupies mostparts of the city. It is also seen that negligible hazard zones in afew pockets of the southeast and southwest parts of the Bangalorecity conform to the sites of site class C. Interestingly, zones of siteclass B in the region is seen with low hazard likely due to theemphasized overburden thickness. It is seen that zones of low tomoderate hazard encompass almost all the sites of old lakes (re-claimed land) as well as existing ones with most of the reclaimedlands in the northern parts under moderate hazard. High hazardconforming to that of liquefaction susceptibility is seen in the areawest of the study region. However, very shallow basement depths(�5 m) is a likely cause that all the areas having factor of safetyagainst liquefaction (FSL) less than or equal to one do not come un-der high hazard in the final hazard map. The patches of maximumhazard are seen at the central region, which can be attributed tothe overburden thickness.

The present analysis accomplishes a groundwork assessment ofthe site specific hazard laying out a framework for higher orderseismic microzonation (1:5000). Period specific analyses basedon spectral accelerations and non-linear site response analysisare also envisage for the future studies to address building typolog-ical ward-wise distribution in the city.

Acknowledgements

The valuable comments and suggestions of the editor and thetwo anonymous reviewers greatly helped in improving the manu-script with enhanced scientific exposition. The present study hasits foundation to the ‘‘Expert Group Meeting”, Bangalore Microzo-nation, March 21, 2008, Indian Institute of Science (IISc) Bangalore,India that comprises of resource persons from IISc Bangalore, IITKharagpur, Geological Survey of India (GSI), and Ministry of EarthScience (MoES, formerly Seismology Division, Department of Sci-ence & Technology), Govt. of India.

References

Aki, K., 1988. Local Site Effects on Strong Ground Motion. Earthquake Engineeringand Soil Dynamics II – Recent Advances in Ground Motion Evaluation, June 27–30, Park City, Utah.

Anbazhagan, P., Sitharam, T.G., 2009. Spatial variability of the weathered andengineering bed rock using multichannel analysis of surface wave survey. Pureand Applied Geophysics 166 (3), 409–428.

Anbazhagan, P., Sitharam, T.G., 2008a. Mapping of average shear wave velocity forBangalore region: a case study. Journal of Environmental and EngineeringGeophysics 13 (2), 69–84.

Anbazhagan, P., Sitharam, T.G., 2008b. Seismic microzonation of Bangalore. Journalof Earth System Science 117 (S2), 833–852.

Anbazhagan, P., Sitharam, T.G., 2008c. Site characterization and site responsestudies using shear wave velocity. Journal of Seismology and EarthquakeEngineering 10 (2), 53–67.

Anbazhagan, P., Sitharam, T.G., Vipin, K.S., 2009. Site classification and estimation ofsurface level seismic hazard using geophysical data and probabilistic approach.Journal of Applied Geophysics 68 (2), 219–230.

Anbazhagan, P., Vinod, J.S., Sitharam, T.G., 2008. Probabilistic seismic hazardanalysis for Bangalore. Natural Hazards 48, 145–166.

Bansal, B.K., Gupta, S., 1998. A glance through the seismicity of peninsular India.Journal of the Geological Society of India 52, 67–80.

BIS, 2002. IS1893-2002 (Part 1): Indian Standard Criteria for Earthquake ResistantDesign of Structure Part 1– Resistant Provisions and Buildings, Bureau of IndianStandards, New Delhi.

Boore, D.M., 1983. Stochastic simulation of high frequency ground motions basedon seismological models of the radiated spectra. Bulletin of the SeismologicalSociety of America 73, 1865–1894.

Building Seismic Safety Council, 2001. NEHRP recommended provisions for seismicregulations for new buildings and other structures. In: 2000 Ed., Part 1:Provisions, Prepared by the Building Seismic Safety Council for the FederalEmergency Management Agency (Report FEMA 368), Washington, DC.

Cetin, K.O., Seed, R.B., Kiureghian, A.D., Tokimatsu, K., Harder Jr., L.F., Kayen, R.E.,Moss, R.E.S., 2004. Standard penetration test-based probabilistic anddeterministic assessment of seismic soil liquefaction potential. Journal ofGeotechnical and Geoenvironmental Engineering 12, 1314–1340.

Dasgupta, S., Pande, P., Ganguly, D., Iqbal, Z., Sanyal, K., Venaktraman, N.V.,Dasgupta, S., Sural, B., Harendranath, L., Mazumadar, K., Sanyal, S., Roy, A., Das,L.K., Misra, P.S., Gupta, H., 2000. Seismotectonic Atlas of India and its Environs,Geological Survey of India, Calcutta, India.

Dobry, R., Borcherdt, R.D., Crouse, C.B., Idriss, I.M., Joyner, W.B., Martin, G.R., Power,M.S., Rinne, E.E., Seed, R.B., 2000. New site coefficients and site classificationsystem used in recent building seismic code provisions. Earthquake Spectra 16,41–67.

DST, 2007. Seismic microzonation atlas of Guwahati region, Seismology Division,Department of Science & Technology, Govt. of India, New Delhi.

Field, E.H., Jacob, K.H., Hough, S.E., 1992. Earthquake site response estimation: aweak motion case study. Bulletin of the Seismological Society of America 82,2283–2307.

Ganesha Raj, K., Nijagunappa, R., 2004. Major lineaments of Karnataka State andtheir relation to seismicity: remote sensing based analysis. Journal of theGeological Society of India 63, 430–439.

Iyengar, R.N., Raghukanth, S.T.G., 2004. Attenuation of strong ground motion inpeninsular India. Seismological Research Letter 75, 530–540.

Kanli, A.I., Tildy, P., Pronay, Z., Pinar, A., Hemann, L., 2006. Vs30 mapping and soilclassification for seismic site effect evaluation in Dinar region, SW Turkey.Geophysical Journal International 165, 223–235.

McGuire, R.K., 2001. Deterministic vs. probabilistic earthquake hazard and risks.Soil Dynamics and Earthquake Engineering 21, 377–384.

Mohanty, W.K., Walling, M.Y., Nath, S.K., Pal, I., 2007. First order seismicmicrozonation of Delhi, India using geographic information system (GIS).Natural Hazard 40, 245–260.

Nath, S.K., Thingbaijam, K.K.S., Raj, A., 2008. Earthquake hazard in the NortheastIndia - A seismic microzonation approach with typical case studies from SikkimHimalaya and Guwahati city. Journal of Earth System Science 117, 809–831.

Nath, S.K., Sengupta, P., Sengupta, S., Chakrabarti, A., 2000. Site response estimationusing strong motion network: a step towards microzonation of SikkimHimalayas seismology. Current Science 79, 1316–1326.

Nath, S.K., Raj, A., Thingbaijam, K.K.S., Kumar, A., 2009. Ground motion synthesisand seismic scenario in Guwahati City – a stochastic approach. SeismologicalResearch Letters 80 (2), 233–242.

Nath, S.K., Thingbaijam, K.K.S., Raj, A., Shukla, K., Pal, I., Nandy, D.R., Yadav, M.K,Bansal, B.K., Dasgupta, S., Majumdar, K., Kayal, J.R., Shukla, A.K., Deb, S.K.,Pathak, J., Hazarika, P.J., Paul, D.K., 2007. Seismic scenario of Guwahati City. In:Proceedings of the International Workshop on Earthquake Hazard andMitigations, 7–8 December, pp. 210–218.

Ordonez, G.A., 2004. SHAKE2000 A Computer Program for the 1D Analysis ofGeotechnical Earthquake Engineering Problems. User’s Manual, 306pp.

Pal, I., Nath, S.K., Shukla, K., Pal, D.K., Raj, A., Thingbaijam, K.K.S., Bansal, B.K., 2008.Earthquake hazard zonation of Sikkim Himalaya using a GIS platform. NaturalHazards 45, 333–377.

Rastogi, B.K., 1992. Seismotectonics inferred from earthquakes and Earthquakesequences in India during the 1980s. Current Science 62, 101–108.

Saaty, T.L., 1980. The Analytic Hierarchy Process. McGraw-Hill, New York, NY.Saaty, T.L., 1990. Multicriteria Decision Making: The Analytic Hierarchy Process.

RWS Publications, Pittsburgh, PA.Seed, H.B., Idriss, I.M., 1971. Simplified procedure for evaluating soil liquefaction

potential. Journal of Soil Mechanics and Foundation Division, American Societyof Civil Engineers 97, 1249–1274.

Seed, H.B., Idriss, I.M., Arango, I., 1983. Evaluation of liquefaction potential usingfield performance data. Journal Geotechnical Engineering 109, 458–482.

Seed, H.B., Idriss, I.M., 1982. Ground Motions and Soil Liquefaction DuringEarthquakes, Monogr. 5. Earthquake Engineering Research Institute,University of California, Berkeley.

Singh, S.K., Ordaz, M., Dattatrayam, R.S., Gupta, H.K., 1999. A spectral analysis of the21 May 1997, Jabalpur, India, earthquake (MW = 5.8) and estimation of groundmotion from future earthquakes in the Indian shield region. Bulletin of theSeismological Society of America 89 (6), 1620–1630.

Sitharam, T.G., Anbazhagan, P., 2007a. Seismic hazard analysis for Bangalore region.Natural Hazard 40, 261–278.

Sitharam, T.G., Anbazhagan, P., 2007b. Seismic microzonation of Bangalore. In:Proceeding of Workshop on Microzonation. Indian Institute of Science,Bangalore, pp. 44–75.

Sitharam, T.G., Anbazhagan, P., Ganesha Raj, K., 2006. Use of remote sensing andseismotectonic parameters for seismic hazard analysis of Bangalore. NaturalHazards Earth System Science 6, 927–939.



Sitharam, T.G., Anbazhagan, P., Mahesh, G.U., 2007. Liquefaction hazard mappingusing SPT data. Indian Geotechnical Journal 37 (3), 210–226.

Sitharam, T.G., Samui, Pijush., Anbazhagan, P., 2008. Spatial variability of rock depthin Bangalore using geostatistical neural network and support vector machinemodels. Journal of Geotechnical and Geological Engineering 26, 503–517.

Sitharam, T.G., Anbazhagan, P., 2008. Seismic Microzonation: Principles, Practicesand Experiments. EJGE Special Volume Bouquet 08, online, p. 61. <http://www.ejge.com/Bouquet08/Preface.htm>.

TC4-ISSMGE, 1999. Manual for Zonation on Seismic Geotechnical Hazard, Revisededition, Technical Committee for Earthquake Geotechnical Engineering (TC4) ofthe International Society of Soil Mechanics and Geotechnical Engineering(ISSMGE), 209p.

Valdiya, K.S., 1998. Late quaternary movements and landscape rejuvenation insouth-eastern Karnataka and adjoining Tamil Nadu in southern India shield.Journal of the Geological Society of India 51, 139–166.

Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships among magnitude,rupture length, rupture width, rupture area, and surface displacement. Bulletinof the Seismological Society of America 84, 975–1002.

Youd, T.L., Idriss, I.M., Andrus, R.D., Arango, I., Castro, G., Christian, J.T., Dobry, R.,Liam Finn, W.D., Harder Jr., L.H., Hynes, M.E., Ishihara, K., Koester, J.P., Liao,S.S.C., Marcuson, W.F., Marting, G.R., Mitchell, J.K., Moriwaki, Y., Power, M.S.,Robertson, P.K., Seed, R.B., Stokoe, K.H.I.I., 2001. Liquefaction resistance of soils:summary from the 1996 NCEER. NCEER/NSF workshops on evaluation ofliquefaction resistance of soils. Journal of Geotechnical and GeoenvironmentalEngineering 127, 817–833.

Youd, T.L., Idriss, I.M., 2001. Liquefaction resistance of soils: summary report fromthe 1996 NCEER and 1998 NCEER/NSF workshops on evaluation of liquefactionresistance of soils. Journal of Geotechnical and Geoenvironmental Engineering127, 297–313.


Documents

Author's personal copy - civil.iisc.ac.incivil.iisc.ac.in/~anbazhagan/pdf/Anbu IISc J-17.pdf · batore earthquake ofMW =6.2 occurred near the Bhavani fault that shook the city. In